NUS-I2R: Learning a Combined System for Entity Linking

Abstract

In this paper, we report the joint participation of NUS and I2R team in Knowledge Base Popula-tion at Text analysis conference 2010. For Entity Linking, we analyze IR approaches and SVM classification in the disambiguation stage and develop a supervised learner for combining these approaches. The combined system performs bet-ter than the individual components and achieves results much better than the median. Further-more, according to our error analysis, quite some errors are caused due to the different Wikipedia version is used, which hinder our system to show significant better performance.